X-Git-Url: https://git.whamcloud.com/?a=blobdiff_plain;f=lustre%2Finclude%2Flustre_net.h;h=678e4820da5f46362c8852d6cc3107fd0a44ede9;hb=816c9f01d5b44e05437b890aab4ef50edc02230f;hp=32743349c397b2010ec4acc57d278beb090b6616;hpb=e02cb40761ff8aae3df76c4210a345420b6d4ba1;p=fs%2Flustre-release.git diff --git a/lustre/include/lustre_net.h b/lustre/include/lustre_net.h index 3274334..678e482 100644 --- a/lustre/include/lustre_net.h +++ b/lustre/include/lustre_net.h @@ -51,11 +51,6 @@ * @{ */ #include -#ifdef HAVE_RHASHTABLE_LOOKUP_GET_INSERT_FAST -#include -#else -#include -#endif #include #include #include @@ -81,7 +76,7 @@ * value. The client is free to limit the actual RPC size for any bulk * transfer via cl_max_pages_per_rpc to some non-power-of-two value. * NOTE: This is limited to 16 (=64GB RPCs) by IOOBJ_MAX_BRW_BITS. */ -#define PTLRPC_BULK_OPS_BITS 4 +#define PTLRPC_BULK_OPS_BITS 6 #if PTLRPC_BULK_OPS_BITS > 16 #error "More than 65536 BRW RPCs not allowed by IOOBJ_MAX_BRW_BITS." #endif @@ -472,52 +467,58 @@ /** * OST_IO_MAXREQSIZE ~= - * lustre_msg + ptlrpc_body + obdo + obd_ioobj + - * DT_MAX_BRW_PAGES * niobuf_remote + * lustre_msg + ptlrpc_body + obdo + obd_ioobj + + * DT_MAX_BRW_PAGES * niobuf_remote * * - single object with 16 pages is 512 bytes - * - OST_IO_MAXREQSIZE must be at least 1 page of cookies plus some spillover + * - OST_IO_MAXREQSIZE must be at least 1 niobuf per page of data * - Must be a multiple of 1024 + * - should allow a reasonably large SHORT_IO_BYTES size (64KB) */ -#define _OST_MAXREQSIZE_BASE (sizeof(struct lustre_msg) + \ - sizeof(struct ptlrpc_body) + \ - sizeof(struct obdo) + \ - sizeof(struct obd_ioobj) + \ - sizeof(struct niobuf_remote)) -#define _OST_MAXREQSIZE_SUM (_OST_MAXREQSIZE_BASE + \ - sizeof(struct niobuf_remote) * \ - (DT_MAX_BRW_PAGES - 1)) +#define _OST_MAXREQSIZE_BASE ((unsigned long)(sizeof(struct lustre_msg) + \ + /* lm_buflens */ sizeof(__u32) * 4 + \ + sizeof(struct ptlrpc_body) + \ + sizeof(struct obdo) + \ + sizeof(struct obd_ioobj) + \ + sizeof(struct niobuf_remote))) +#define _OST_MAXREQSIZE_SUM ((unsigned long)(_OST_MAXREQSIZE_BASE + \ + sizeof(struct niobuf_remote) * \ + DT_MAX_BRW_PAGES)) /** * FIEMAP request can be 4K+ for now */ -#define OST_MAXREQSIZE (16 * 1024) -#define OST_IO_MAXREQSIZE max_t(int, OST_MAXREQSIZE, \ - (((_OST_MAXREQSIZE_SUM - 1) | (1024 - 1)) + 1)) +#define OST_MAXREQSIZE (16UL * 1024UL) +#define OST_IO_MAXREQSIZE max(OST_MAXREQSIZE, \ + ((_OST_MAXREQSIZE_SUM - 1) | \ + (1024UL - 1)) + 1) /* Safe estimate of free space in standard RPC, provides upper limit for # of * bytes of i/o to pack in RPC (skipping bulk transfer). */ -#define OST_SHORT_IO_SPACE (OST_IO_MAXREQSIZE - _OST_MAXREQSIZE_BASE) +#define OST_MAX_SHORT_IO_BYTES ((OST_IO_MAXREQSIZE - _OST_MAXREQSIZE_BASE) & \ + PAGE_MASK) /* Actual size used for short i/o buffer. Calculation means this: * At least one page (for large PAGE_SIZE), or 16 KiB, but not more * than the available space aligned to a page boundary. */ -#define OBD_MAX_SHORT_IO_BYTES (min(max(PAGE_SIZE, 16UL * 1024UL), \ - OST_SHORT_IO_SPACE & PAGE_MASK)) +#define OBD_DEF_SHORT_IO_BYTES min(max(PAGE_SIZE, 16UL * 1024UL), \ + OST_MAX_SHORT_IO_BYTES) #define OST_MAXREPSIZE (9 * 1024) #define OST_IO_MAXREPSIZE OST_MAXREPSIZE #define OST_NBUFS 64 /** OST_BUFSIZE = max_reqsize + max sptlrpc payload size */ -#define OST_BUFSIZE max_t(int, OST_MAXREQSIZE + 1024, 16 * 1024) +#define OST_BUFSIZE max_t(int, OST_MAXREQSIZE + 1024, 32 * 1024) /** * OST_IO_MAXREQSIZE is 18K, giving extra 46K can increase buffer utilization * rate of request buffer, please check comment of MDS_LOV_BUFSIZE for details. */ #define OST_IO_BUFSIZE max_t(int, OST_IO_MAXREQSIZE + 1024, 64 * 1024) - -/* Macro to hide a typecast. */ -#define ptlrpc_req_async_args(req) ((void *)&req->rq_async_args) +/* Macro to hide a typecast and BUILD_BUG. */ +#define ptlrpc_req_async_args(_var, req) ({ \ + BUILD_BUG_ON(sizeof(*_var) > sizeof(req->rq_async_args)); \ + (typeof(_var))&req->rq_async_args; \ + }) struct ptlrpc_replay_async_args { int praa_old_state; @@ -529,7 +530,7 @@ struct ptlrpc_replay_async_args { */ struct ptlrpc_connection { /** linkage for connections hash table */ - struct rhash_head c_hash; + struct hlist_node c_hash; /** Our own lnet nid for this connection */ lnet_nid_t c_self; /** Remote side nid for this connection */ @@ -547,7 +548,7 @@ struct ptlrpc_client { /** What portal do we expect replies on */ __u32 cli_reply_portal; /** Name of the client */ - char *cli_name; + const char *cli_name; }; /** state flags of requests */ @@ -558,15 +559,15 @@ struct ptlrpc_client { #define REQ_MAX_ACK_LOCKS 8 union ptlrpc_async_args { - /** - * Scratchpad for passing args to completion interpreter. Users - * cast to the struct of their choosing, and CLASSERT that this is - * big enough. For _tons_ of context, OBD_ALLOC a struct and store - * a pointer to it here. The pointer_arg ensures this struct is at - * least big enough for that. - */ - void *pointer_arg[11]; - __u64 space[7]; + /** + * Scratchpad for passing args to completion interpreter. Users + * cast to the struct of their choosing, and BUILD_BUG_ON that this is + * big enough. For _tons_ of context, OBD_ALLOC a struct and store + * a pointer to it here. The pointer_arg ensures this struct is at + * least big enough for that. + */ + void *pointer_arg[11]; + __u64 space[7]; }; struct ptlrpc_request_set; @@ -1001,7 +1002,8 @@ struct ptlrpc_request { unsigned int rq_hp:1, /**< high priority RPC */ rq_at_linked:1, /**< link into service's srv_at_array */ - rq_packed_final:1; /**< packed final reply */ + rq_packed_final:1, /**< packed final reply */ + rq_obsolete:1; /* aborted by a signal on a client */ /** @} */ /** one of RQ_PHASE_* */ @@ -1053,6 +1055,13 @@ struct ptlrpc_request { /** description of flavors for client & server */ struct sptlrpc_flavor rq_flvr; + /** + * SELinux policy info at the time of the request + * sepol string format is: + * ::: + */ + char rq_sepol[LUSTRE_NODEMAP_SEPOL_LENGTH + 1]; + /* client/server security flags */ unsigned int rq_ctx_init:1, /* context initiation */ @@ -1129,13 +1138,14 @@ struct ptlrpc_request { static inline int ptlrpc_req_interpret(const struct lu_env *env, struct ptlrpc_request *req, int rc) { - if (req->rq_interpret_reply != NULL) { - req->rq_status = req->rq_interpret_reply(env, req, - &req->rq_async_args, - rc); - return req->rq_status; - } - return rc; + if (req->rq_interpret_reply != NULL) { + req->rq_status = req->rq_interpret_reply(env, req, + &req->rq_async_args, + rc); + return req->rq_status; + } + + return rc; } /** \addtogroup nrs @@ -1273,12 +1283,15 @@ ptlrpc_rqphase2str(struct ptlrpc_request *req) FLAG(req->rq_err, "E"), FLAG(req->rq_net_err, "e"), \ FLAG(req->rq_timedout, "X") /* eXpired */, FLAG(req->rq_resend, "S"), \ FLAG(req->rq_restart, "T"), FLAG(req->rq_replay, "P"), \ - FLAG(req->rq_no_resend, "N"), \ + FLAG(req->rq_no_resend, "N"), FLAG(req->rq_no_reply, "n"), \ FLAG(req->rq_waiting, "W"), \ FLAG(req->rq_wait_ctx, "C"), FLAG(req->rq_hp, "H"), \ - FLAG(req->rq_committed, "M") + FLAG(req->rq_committed, "M"), \ + FLAG(req->rq_req_unlinked, "Q"), \ + FLAG(req->rq_reply_unlinked, "U"), \ + FLAG(req->rq_receiving_reply, "r") -#define REQ_FLAGS_FMT "%s:%s%s%s%s%s%s%s%s%s%s%s%s%s" +#define REQ_FLAGS_FMT "%s:%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s" void _debug_req(struct ptlrpc_request *req, struct libcfs_debug_msg_data *data, const char *fmt, ...) @@ -1315,29 +1328,11 @@ do { \ } while (0) /** @} */ -/** - * Structure that defines a single page of a bulk transfer - */ -struct ptlrpc_bulk_page { - /** Linkage to list of pages in a bulk */ - struct list_head bp_link; - /** - * Number of bytes in a page to transfer starting from \a bp_pageoffset - */ - int bp_buflen; - /** offset within a page */ - int bp_pageoffset; - /** The page itself */ - struct page *bp_page; -}; - enum ptlrpc_bulk_op_type { PTLRPC_BULK_OP_ACTIVE = 0x00000001, PTLRPC_BULK_OP_PASSIVE = 0x00000002, PTLRPC_BULK_OP_PUT = 0x00000004, PTLRPC_BULK_OP_GET = 0x00000008, - PTLRPC_BULK_BUF_KVEC = 0x00000010, - PTLRPC_BULK_BUF_KIOV = 0x00000020, PTLRPC_BULK_GET_SOURCE = PTLRPC_BULK_OP_PASSIVE | PTLRPC_BULK_OP_GET, PTLRPC_BULK_PUT_SINK = PTLRPC_BULK_OP_PASSIVE | PTLRPC_BULK_OP_PUT, PTLRPC_BULK_GET_SINK = PTLRPC_BULK_OP_ACTIVE | PTLRPC_BULK_OP_GET, @@ -1369,18 +1364,6 @@ static inline bool ptlrpc_is_bulk_put_source(enum ptlrpc_bulk_op_type type) return (type & PTLRPC_BULK_PUT_SOURCE) == PTLRPC_BULK_PUT_SOURCE; } -static inline bool ptlrpc_is_bulk_desc_kvec(enum ptlrpc_bulk_op_type type) -{ - return ((type & PTLRPC_BULK_BUF_KVEC) | (type & PTLRPC_BULK_BUF_KIOV)) - == PTLRPC_BULK_BUF_KVEC; -} - -static inline bool ptlrpc_is_bulk_desc_kiov(enum ptlrpc_bulk_op_type type) -{ - return ((type & PTLRPC_BULK_BUF_KVEC) | (type & PTLRPC_BULK_BUF_KIOV)) - == PTLRPC_BULK_BUF_KIOV; -} - static inline bool ptlrpc_is_bulk_op_active(enum ptlrpc_bulk_op_type type) { return ((type & PTLRPC_BULK_OP_ACTIVE) | @@ -1420,7 +1403,6 @@ struct ptlrpc_bulk_frag_ops { extern const struct ptlrpc_bulk_frag_ops ptlrpc_bulk_kiov_pin_ops; extern const struct ptlrpc_bulk_frag_ops ptlrpc_bulk_kiov_nopin_ops; -extern const struct ptlrpc_bulk_frag_ops ptlrpc_bulk_kvec_ops; /* * Definition of bulk descriptor. @@ -1439,8 +1421,6 @@ struct ptlrpc_bulk_desc { unsigned long bd_registered:1; /** For serialization with callback */ spinlock_t bd_lock; - /** Import generation when request for this bulk was sent */ - int bd_import_generation; /** {put,get}{source,sink}{kvec,kiov} */ enum ptlrpc_bulk_op_type bd_type; /** LNet portal for this bulk */ @@ -1451,7 +1431,7 @@ struct ptlrpc_bulk_desc { struct obd_import *bd_import; /** Back pointer to the request */ struct ptlrpc_request *bd_req; - struct ptlrpc_bulk_frag_ops *bd_frag_ops; + const struct ptlrpc_bulk_frag_ops *bd_frag_ops; wait_queue_head_t bd_waitq; /* server side only WQ */ int bd_iov_count; /* # entries in bd_iov */ int bd_max_iov; /* allocated size of bd_iov */ @@ -1467,40 +1447,17 @@ struct ptlrpc_bulk_desc { /** array of associated MDs */ struct lnet_handle_md bd_mds[PTLRPC_BULK_OPS_COUNT]; - union { - struct { - /* - * encrypt iov, size is either 0 or bd_iov_count. - */ - lnet_kiov_t *bd_enc_vec; - lnet_kiov_t *bd_vec; - } bd_kiov; - - struct { - struct kvec *bd_enc_kvec; - struct kvec *bd_kvec; - } bd_kvec; - } bd_u; - + /* encrypted iov, size is either 0 or bd_iov_count. */ + lnet_kiov_t *bd_enc_vec; + lnet_kiov_t *bd_vec; }; -#define GET_KIOV(desc) ((desc)->bd_u.bd_kiov.bd_vec) -#define BD_GET_KIOV(desc, i) ((desc)->bd_u.bd_kiov.bd_vec[i]) -#define GET_ENC_KIOV(desc) ((desc)->bd_u.bd_kiov.bd_enc_vec) -#define BD_GET_ENC_KIOV(desc, i) ((desc)->bd_u.bd_kiov.bd_enc_vec[i]) -#define GET_KVEC(desc) ((desc)->bd_u.bd_kvec.bd_kvec) -#define BD_GET_KVEC(desc, i) ((desc)->bd_u.bd_kvec.bd_kvec[i]) -#define GET_ENC_KVEC(desc) ((desc)->bd_u.bd_kvec.bd_enc_kvec) -#define BD_GET_ENC_KVEC(desc, i) ((desc)->bd_u.bd_kvec.bd_enc_kvec[i]) - enum { SVC_INIT = 0, SVC_STOPPED = 1 << 0, SVC_STOPPING = 1 << 1, SVC_STARTING = 1 << 2, SVC_RUNNING = 1 << 3, - SVC_EVENT = 1 << 4, - SVC_SIGNAL = 1 << 5, }; #define PTLRPC_THR_NAME_LEN 32 @@ -1509,7 +1466,7 @@ enum { */ struct ptlrpc_thread { /** - * List of active threads in svc->srv_threads + * List of active threads in svcpt->scp_threads */ struct list_head t_link; /** @@ -1522,13 +1479,15 @@ struct ptlrpc_thread { */ unsigned int t_id; /** - * service thread pid + * service thread */ + struct task_struct *t_task; pid_t t_pid; + ktime_t t_touched; /** * put watchdog in the structure per thread b=14840 */ - struct lc_watchdog *t_watchdog; + struct delayed_work t_watchdog; /** * the svc this thread belonged to b=18582 */ @@ -1563,16 +1522,6 @@ static inline int thread_is_running(struct ptlrpc_thread *thread) return !!(thread->t_flags & SVC_RUNNING); } -static inline int thread_is_event(struct ptlrpc_thread *thread) -{ - return !!(thread->t_flags & SVC_EVENT); -} - -static inline int thread_is_signal(struct ptlrpc_thread *thread) -{ - return !!(thread->t_flags & SVC_SIGNAL); -} - static inline void thread_clear_flags(struct ptlrpc_thread *thread, __u32 flags) { thread->t_flags &= ~flags; @@ -1682,8 +1631,6 @@ struct ptlrpc_service { char *srv_name; /** only statically allocated strings here; we don't clean them */ char *srv_thread_name; - /** service thread list */ - struct list_head srv_threads; /** threads # should be created for each partition on initializing */ int srv_nthrs_cpt_init; /** limit of threads number for each partition */ @@ -1715,14 +1662,16 @@ struct ptlrpc_service { int srv_watchdog_factor; /** under unregister_service */ unsigned srv_is_stopping:1; + /** Whether or not to restrict service threads to CPUs in this CPT */ + unsigned srv_cpt_bind:1; /** max # request buffers */ int srv_nrqbds_max; /** max # request buffers in history per partition */ int srv_hist_nrqbds_cpt_max; - /** number of CPTs this service bound on */ + /** number of CPTs this service associated with */ int srv_ncpts; - /** CPTs array this service bound on */ + /** CPTs array this service associated with */ __u32 *srv_cpts; /** 2^srv_cptab_bits >= cfs_cpt_numbert(srv_cptable) */ int srv_cpt_bits; @@ -1765,8 +1714,6 @@ struct ptlrpc_service_part { int scp_thr_nextid; /** # of starting threads */ int scp_nthrs_starting; - /** # of stopping threads, reserved for shrinking threads */ - int scp_nthrs_stopping; /** # running threads */ int scp_nthrs_running; /** service threads list */ @@ -1990,7 +1937,7 @@ static inline bool nrs_policy_compat_one(const struct ptlrpc_service *svc, /** @} nrs */ /* ptlrpc/events.c */ -extern struct lnet_handle_eq ptlrpc_eq_h; +extern struct lnet_eq *ptlrpc_eq; extern int ptlrpc_uuid_to_peer(struct obd_uuid *uuid, struct lnet_process_id *peer, lnet_nid_t *self); /** @@ -2113,7 +2060,7 @@ int ptlrpc_register_rqbd(struct ptlrpc_request_buffer_desc *rqbd); */ void ptlrpc_request_committed(struct ptlrpc_request *req, int force); -void ptlrpc_init_client(int req_portal, int rep_portal, char *name, +void ptlrpc_init_client(int req_portal, int rep_portal, const char *name, struct ptlrpc_client *); void ptlrpc_cleanup_client(struct obd_import *imp); struct ptlrpc_connection *ptlrpc_uuid_to_connection(struct obd_uuid *uuid, @@ -2131,9 +2078,9 @@ struct ptlrpc_request_set *ptlrpc_prep_fcset(int max, set_producer_func func, void *arg); int ptlrpc_check_set(const struct lu_env *env, struct ptlrpc_request_set *set); int ptlrpc_set_wait(const struct lu_env *env, struct ptlrpc_request_set *); -void ptlrpc_mark_interrupted(struct ptlrpc_request *req); void ptlrpc_set_destroy(struct ptlrpc_request_set *); void ptlrpc_set_add_req(struct ptlrpc_request_set *, struct ptlrpc_request *); +#define PTLRPCD_SET ((struct ptlrpc_request_set *)1) void ptlrpc_free_rq_pool(struct ptlrpc_request_pool *pool); int ptlrpc_add_rqs_to_pool(struct ptlrpc_request_pool *pool, int num_rq); @@ -2167,35 +2114,12 @@ struct ptlrpc_bulk_desc *ptlrpc_prep_bulk_imp(struct ptlrpc_request *req, const struct ptlrpc_bulk_frag_ops *ops); -int ptlrpc_prep_bulk_frag(struct ptlrpc_bulk_desc *desc, - void *frag, int len); void __ptlrpc_prep_bulk_page(struct ptlrpc_bulk_desc *desc, struct page *page, int pageoffset, int len, int pin); -static inline void ptlrpc_prep_bulk_page_pin(struct ptlrpc_bulk_desc *desc, - struct page *page, int pageoffset, - int len) -{ - __ptlrpc_prep_bulk_page(desc, page, pageoffset, len, 1); -} - -static inline void ptlrpc_prep_bulk_page_nopin(struct ptlrpc_bulk_desc *desc, - struct page *page, int pageoffset, - int len) -{ - __ptlrpc_prep_bulk_page(desc, page, pageoffset, len, 0); -} void ptlrpc_free_bulk(struct ptlrpc_bulk_desc *bulk); -static inline void ptlrpc_release_bulk_page_pin(struct ptlrpc_bulk_desc *desc) -{ - int i; - - for (i = 0; i < desc->bd_iov_count ; i++) - put_page(BD_GET_KIOV(desc, i).kiov_page); -} - static inline void ptlrpc_release_bulk_noop(struct ptlrpc_bulk_desc *desc) { } @@ -2205,6 +2129,8 @@ void ptlrpc_retain_replayable_request(struct ptlrpc_request *req, __u64 ptlrpc_next_xid(void); __u64 ptlrpc_sample_next_xid(void); __u64 ptlrpc_req_xid(struct ptlrpc_request *request); +void ptlrpc_get_mod_rpc_slot(struct ptlrpc_request *req); +void ptlrpc_put_mod_rpc_slot(struct ptlrpc_request *req); /* Set of routines to run a function in ptlrpcd context */ void *ptlrpcd_alloc_work(struct obd_import *imp, @@ -2247,8 +2173,8 @@ struct ptlrpc_service_thr_conf { /* user specified threads number, it will be validated due to * other members of this structure. */ unsigned int tc_nthrs_user; - /* set NUMA node affinity for service threads */ - unsigned int tc_cpu_affinity; + /* bind service threads to only CPUs in their associated CPT */ + unsigned int tc_cpu_bind; /* Tags for lu_context associated with service thread */ __u32 tc_ctx_tags; }; @@ -2257,6 +2183,8 @@ struct ptlrpc_service_cpt_conf { struct cfs_cpt_table *cc_cptable; /* string pattern to describe CPTs for a service */ char *cc_pattern; + /* whether or not to have per-CPT service partitions */ + bool cc_affinity; }; struct ptlrpc_service_conf { @@ -2313,6 +2241,7 @@ void ptlrpc_hr_fini(void); * @{ */ int ptlrpc_connect_import(struct obd_import *imp); +int ptlrpc_connect_import_locked(struct obd_import *imp); int ptlrpc_init_import(struct obd_import *imp); int ptlrpc_disconnect_import(struct obd_import *imp, int noclose); int ptlrpc_disconnect_and_idle_import(struct obd_import *imp); @@ -2329,6 +2258,12 @@ int ptlrpc_reconnect_import(struct obd_import *imp); * * @{ */ +#define PTLRPC_MAX_BUFCOUNT \ + (sizeof(((struct ptlrpc_request *)0)->rq_req_swab_mask) * 8) +#define MD_MAX_BUFLEN (MDS_REG_MAXREQSIZE > OUT_MAXREQSIZE ? \ + MDS_REG_MAXREQSIZE : OUT_MAXREQSIZE) +#define PTLRPC_MAX_BUFLEN (OST_IO_MAXREQSIZE > MD_MAX_BUFLEN ? \ + OST_IO_MAXREQSIZE : MD_MAX_BUFLEN) bool ptlrpc_buf_need_swab(struct ptlrpc_request *req, const int inout, __u32 index); void ptlrpc_buf_set_swabbed(struct ptlrpc_request *req, const int inout, @@ -2350,6 +2285,7 @@ int lustre_pack_reply_flags(struct ptlrpc_request *, int count, __u32 *lens, char **bufs, int flags); int lustre_shrink_msg(struct lustre_msg *msg, int segment, unsigned int newlen, int move_data); +int lustre_grow_msg(struct lustre_msg *msg, int segment, unsigned int newlen); void lustre_free_reply_state(struct ptlrpc_reply_state *rs); int __lustre_unpack_msg(struct lustre_msg *m, int len); __u32 lustre_msg_hdr_size(__u32 magic, __u32 count); @@ -2480,7 +2416,7 @@ ptlrpc_rqphase_move(struct ptlrpc_request *req, enum rq_phase new_phase) atomic_dec(&req->rq_import->imp_unregistering); } - DEBUG_REQ(D_INFO, req, "move req \"%s\" -> \"%s\"", + DEBUG_REQ(D_INFO, req, "move request phase from %s to %s", ptlrpc_rqphase2str(req), ptlrpc_phase2str(new_phase)); req->rq_phase = new_phase;